eCite Digital Repository

Cluster analysis of commercial samples of Bauhinia spp. using HPLC-UV/PDA and MCR-ALS/PCA without peak alignment procedure


Ardila, JA and Funari, CS and Andrade, AM and Cavalheiro, AJ and Carneiro, RL, Cluster analysis of commercial samples of Bauhinia spp. using HPLC-UV/PDA and MCR-ALS/PCA without peak alignment procedure, Phytochemical Analysis, 26, (5) pp. 367-373. ISSN 0958-0344 (2015) [Refereed Article]

Copyright Statement

Copyright 2015 John Wiley & Sons, Ltd.

DOI: doi:10.1002/pca.2571


Introduction - Bauhinia forficata Link. is recognised by the Brazilian Health Ministry as a treatment of hypoglycemia and diabetes. Analytical methods are useful to assess the plant identity due the similarities found in plants from Bauhinia spp. HPLC-UV/PDA in combination with chemometric tools is an alternative widely used and suitable for authentication of plant material, however, the shifts of retention times for similar compounds in different samples is a problem.

Objective - To perform comparisons between the authentic medicinal plant (Bauhinia forficata Link.) and samples commercially available in drugstores claiming to be "Bauhinia spp. to treat diabetes" and to evaluate the performance of multivariate curve resolution alternating least squares (MCR-ALS) associated to principal component analysis (PCA) when compared to pure PCA.

Methodology - HPLC-UV/PDA data obtained from extracts of leaves were evaluated employing a combination of MCR-ALS and PCA, which allowed the use of the full chromatographic and spectrometric information without the need of peak alignment procedures.

Results - The use of MCR-ALS/PCA showed better results than the conventional PCA using only one wavelength. Only two of nine commercial samples presented characteristics similar to the authentic Bauhinia forficata spp., considering the full HPLC-UV/PDA data.

Conclusion - The combination of MCR-ALS and PCA is very useful when applied to a group of samples where a general alignment procedure could not be applied due to the different chromatographic profiles. This work also demonstrates the need of more strict control from the health authorities regarding herbal products available on the market.

Item Details

Item Type:Refereed Article
Keywords:Bauhinia forficata, multivariate, curve resolution, natural products, chromatographic fingerprint
Research Division:Chemical Sciences
Research Group:Analytical chemistry
Research Field:Separation science
Objective Division:Expanding Knowledge
Objective Group:Expanding knowledge
Objective Field:Expanding knowledge in the chemical sciences
UTAS Author:Funari, CS (Dr Cristiano Funari)
ID Code:105812
Year Published:2015
Web of Science® Times Cited:15
Deposited By:Austn Centre for Research in Separation Science
Deposited On:2016-01-14
Last Modified:2016-08-11

Repository Staff Only: item control page